Despite the availability of one pill once daily treatment for HIV, poor adherence with HIV antiretroviral therapy (ART) clearly continues to be widespread. It is also clear that non-infectious comorbidity such as diabetes and cardiovascular disease is increasingly common in persons with HIV, and that medications required for these other conditions further complicate patients'adherence efforts. Unfortunately, clinicians and researchers have yet to develop inexpensive and accurate methods to diagnose medication adherence problems. Objective methods such as electronic data monitoring are expensive and impractical in clinical care, and there are few well-tested, standardized, accurate methods for self report of medication adherence. Our inability to accurately assess adherence by self report is an important problem, because until we can accurately diagnose adherence problems, for all of a patient's medications, it will not be possible to focus interventions to improve adherence on those who most need them. The long term goal of this research is to link the accurate and timely diagnosis of medication adherence problems with customized treatments that are patient and medication specific. The overall objective of this application, which is the next step toward attainment of our long term goal, is to develop a small set of well-tested items that can be used by clinicians and researchers to diagnose medication adherence problems in persons with HIV. Our central hypothesis is that patients usually cannot remember individual dosing events when they self-report, they estimate. We argue that the rigorous cognitive testing that we plan will produce questionnaire items that are congruent with the actual cognitive processes which patient use to recall and report adherence events, and will thus be optimally accurate. The rationale for choosing this approach is that that the application of methods used by psychologists and cognitive scientists to understand how information is encoded and recalled will provide new insights to the science of medication self-report, and provide important new tools for both clinicians and researchers. The proposed research is relevant to the NIH's mission poor adherence with medications for chronic conditions like HIV is a serious, worldwide problems, and because the accurate identification of medication adherence problems is a precondition for efficient and effective intervention. Guided by strong preliminary data, this R01 proposal has three specific aims: 1) Use in depth cognitive testing to develop optimal medication self-report item sets, 2) Conduct psychometric testing, and 3) Conduct construct validity testing. The approach is innovative because rigorous cognitive testing has not yet been used to optimize the quality of adherence self-report. The proposed research is significant because valid, standardized items for the self-report of medication adherence, across all of a patient's medications, are critically needed for both clinical care and research. Although this proposal focuses on persons with HIV disease, because our testing will focus both on ART and on other non-ART medications, our findings should be generalizable to all persons with chronic conditions. PUBLIC HEALTH RELEVANCE: Achieving optimal health outcomes for persons with HIV requires that they adhere not only with antiretroviral medications, but also with medications for non-infectious conditions such as diabetes. We propose to develop and validate a small set of self-report items that can be used to identify persons with adherence problems so that they can be efficiently targeted for interventions. Although this proposal focuses on persons with HIV disease, the items we develop should be usable for adherence assessment for medications for persons with any chronic condition.